TY - GEN
T1 - Bayesian modeling of lexical resources for low-resource settings
AU - Andrews, Nicholas
AU - Dredze, Mark
AU - Van Durme, Benjamin
AU - Eisner, Jason
N1 - Funding Information:
This work was supported by the JHU Human Language Technology Center of Excellence, DARPA LORELEI, and NSF grant IIS-1423276. Thanks to Jay Feldman for early discussions.
Publisher Copyright:
© 2017 Association for Computational Linguistics.
PY - 2017
Y1 - 2017
N2 - Lexical resources such as dictionaries and gazetteers are often used as auxiliary data for tasks such as part-of-speech induction and named-entity recognition. However, discriminative training with lexical features requires annotated data to reliably estimate the lexical feature weights and may result in overfitting the lexical features at the expense of features which generalize better. In this paper, we investigate a more robust approach: we stipulate that the lexicon is the result of an assumed generative process. Practically, this means that we may treat the lexical resources as observations under the proposed generative model. The lexical resources provide training data for the generative model without requiring separate data to estimate lexical feature weights. We evaluate the proposed approach in two settings: part-of-speech induction and low-resource named-entity recognition.
AB - Lexical resources such as dictionaries and gazetteers are often used as auxiliary data for tasks such as part-of-speech induction and named-entity recognition. However, discriminative training with lexical features requires annotated data to reliably estimate the lexical feature weights and may result in overfitting the lexical features at the expense of features which generalize better. In this paper, we investigate a more robust approach: we stipulate that the lexicon is the result of an assumed generative process. Practically, this means that we may treat the lexical resources as observations under the proposed generative model. The lexical resources provide training data for the generative model without requiring separate data to estimate lexical feature weights. We evaluate the proposed approach in two settings: part-of-speech induction and low-resource named-entity recognition.
UR - http://www.scopus.com/inward/record.url?scp=85040922748&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85040922748&partnerID=8YFLogxK
U2 - 10.18653/v1/P17-1095
DO - 10.18653/v1/P17-1095
M3 - Conference contribution
AN - SCOPUS:85040922748
T3 - ACL 2017 - 55th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
SP - 1029
EP - 1039
BT - ACL 2017 - 55th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
PB - Association for Computational Linguistics (ACL)
T2 - 55th Annual Meeting of the Association for Computational Linguistics, ACL 2017
Y2 - 30 July 2017 through 4 August 2017
ER -